StratifiedSampling

Stratified sampling step with multiple allocation strategies for class imbalance, causal analysis, and variance optimization

SageMaker step type	Processing
Node type	internal (consumes upstream, produces downstream)
Container entry point	stratified_sampling.py
Interface file	steps/interfaces/stratified_sampling.step.yaml

Compute

Compute kind	sklearn

Functionality

Stratified sampling with four allocation strategies (balanced, proportional_min, optimal, external_proportional). Handles class imbalance correction, causal analysis, and variance optimization with per-split diagnostics.

Inputs (dependencies)

Input	Type	Required	Compatible producers
input_data	processing_output	yes	TabularPreprocessing, ProcessingStep

Outputs

Output	Type
processed_data	processing_output

Consumers (downstream steps)

Steps that declare this step as a compatible input source:

• BedrockBatchProcessing

• BedrockProcessing

• CurrencyConversion

• DataUploading

• EdxUploading

• FeatureSelection

• GraphSubgraphExtraction

• LabelRulesetExecution

• LightGBMMTTraining

• LightGBMTraining

• MissingValueImputation

• PseudoLabelMerge

• PyTorchTraining

• TSATabularPreprocessing

• TabularPreprocessing

• XGBoostTraining

• XgboostMtTraining

Framework requirements

Package	Version
pandas	>=1.3.0

---

← Back to the Step Catalog